cs.RO（2026-09-25）

📊 共 37 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (31 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (3) 支柱九：具身大模型 (Embodied Foundation Models) (2) 支柱三：空间感知与语义 (Perception & Semantics) (1 🔗1)

🔬 支柱一：机器人控制 (Robot Control) (31 篇)

#题目一句话要点标签🔗⭐
1 Generate, Track, Improve: Perceptive Multi-Skill Humanoid Locomotion with RL-Fine-Tuned Motion Generators 提出双层运动控制架构以解决人形机器人多技能动态行走问题 humanoid humanoid locomotion locomotion ✅
2 Praxis: Distilling Physical Interaction Priors from Egocentric Videos for Generalizable Whole-Body Manipulation 提出Praxis框架以解决移动人形机器人操控中的数据不足问题 humanoid manipulation whole-body manipulation
3 FRAM: Trajectory-Guided Visual Feature Selection for Compact Language-Conditioned Robot Manipulation 提出FRAM以解决机器人操作中的视觉特征选择问题 manipulation dual-arm vision-language-action
4 CognitiveReality: Robot-Agnostic Semantic Gaussian Mapping with an LLM Agent for Immersive Collaborative VR Teleoperation 提出CognitiveReality以解决机器人语义映射与协作遥操作问题 quadruped teleoperation VR teleoperation
5 InternW0-$Δ$: A World Action Model Bridging Predictive Dynamics and Actions with 20K+ Hours of Open Data 提出InternW0-$Δ$以解决机器人动作生成中的动态预测问题 manipulation world action model world action models ✅
6 VLaRL: Augmenting Vision-Language-Action Models with Simulation-Trained Latent-Conditioned Residual RL 提出VLaRL以解决视觉语言动作模型在真实环境中的执行精度问题 manipulation sim-to-real reinforcement learning
7 Kintsugi-VLA: Turning Failed Robot Rollouts into Recovery Data through Interventional Recoverability 提出Kintsugi-VLA框架以将失败机器人轨迹转化为恢复数据 manipulation teleoperation vision-language-action
8 Fast Plans, Faithful Actions: Closing the Planning-Execution Gap in Hierarchical Vision-Language-Action Models 提出分层视觉语言行动模型以解决规划与执行间的延迟问题 bi-manual dual-arm vision-language-action
9 Enabling a Unified Cross-Domain Representation for Two-Finger Gripper Manipulation via Interaction-Centric Modeling 提出交互中心建模以解决跨领域两指夹持器操作问题 manipulation sim-to-real imitation learning
10 Imp-ACT: Adaptive Impedance Control and Action Chunking with Transformers to Learn Contact-Rich Manipulation from Demonstrations 提出Imp-ACT以解决机器人接触丰富操作中的柔顺性问题 motion tracking manipulation teleoperation
11 PHASE: Compliance-Enabled Tactile Phase Retrieval for Few-Shot Insertion Learning 提出PHASE框架以解决少样本插入学习中的接触丰富任务问题 manipulation policy learning imitation learning
12 Bundled Contact Gradients: Stabilizing Differentiable Simulation for Deployable Dynamic Tasks 提出Bundled Contact Gradients以解决动态任务中的梯度不稳定问题 humanoid Unitree policy learning
13 Impedance Cloning: Learning Equilibrium Point Parameters for Contact-Rich Manipulation 提出阻抗克隆以解决接触丰富操作中的不确定性问题 manipulation teleoperation imitation learning
14 DualManip: Agentic Dynamic Manipulation via Dual-Path Semantic Reasoning and Geometric Adaptation 提出DualManip以解决动态场景下机器人操作的响应性问题 manipulation open-vocabulary open vocabulary ✅
15 Quadruped Obstacle Avoidance and Footstep Planning with Distributed Low-cost Time-of-Flight Sensors 提出分布式ToF传感器以解决四足机器人障碍物规避问题 quadruped locomotion ANYmal
16 Onboard Wind-Preview Model Predictive Control Using Pitot-Static Sensing for Multirotor UAVs 提出基于皮托静压传感器的风预览模型预测控制以提升无人机悬停性能 MPC model predictive control
17 NavGen: Visual Generative Models as a Scalable Data Engine for Embodied 3D Navigation 提出NavGen以解决3D导航数据不足问题 sim-to-real world action model world action models
18 RoboMonitor: Label-Efficient Runtime Monitoring of Robot Task Execution via Predictive Representation Learning 提出RoboMonitor以解决机器人任务执行监控中的标注效率问题 manipulation policy learning representation learning
19 Guiding End-to-End Driving Models with Endpoint-Constrained Trajectory Optimization 提出端点约束优化以解决驾驶模型闭环性能问题 trajectory optimization behavior cloning
20 VisTacAlign: Co-Training Dexterous Policies on Tactile Human and Robot Demonstrations 提出VisTacAlign以解决人机协同训练的挑战 manipulation dexterous manipulation foundation model
21 Precision at Speed: Sample-Efficient Online Model-Based Reinforcement Learning for Hydraulic Excavator Control 提出在线模型基础强化学习框架以提升挖掘机控制精度 model predictive control reinforcement learning
22 dRVG: Quadtree-Guided, Resolution-Complete Online Motion Planning for Polygonal Robots in Unknown Environments 提出dRVG以解决多边形机器人在未知环境中的在线运动规划问题 motion planning
23 Representation-Guided Generation and Integration of Executable Programs for Robot Manipulation 提出RIVET框架以解决机器人操作系统集成问题 manipulation
24 TACTIC: Understanding Tactile Encoders and Conditioning for Contact-rich Robot Manipulation Policies 提出TACTIC以解决触觉编码与融合策略的挑战 manipulation
25 HIRE: History-Conditioned Interaction Reasoning and High-Rate Execution for Visually Aliased Precision Manipulation 提出HIRE以解决视觉相似性导致的精确操作问题 manipulation
26 Vision-Based 6-DoF Grasp Pose Estimation for Robot Cloth Unfolding 提出CeDiRNet-6DoF以解决机器人布料展开问题 manipulation bi-manual
27 ExoLaN: Physics-Consistent Context-Aware Dynamics Learning for Exoskeletons 提出ExoLaN以解决外骨骼控制中的人机交互问题 locomotion latent dynamics
28 See to Reach, Feel to Grasp: Learning A Blind Grasp Reflex for Anthropomorphic Robotic Hands 提出盲抓反射机制以解决无视觉抓取问题 manipulation reinforcement learning
29 Co-design of trajectory and morphology for a vertical jump-climbing robot 提出一种新型跳跃攀爬机器人以解决垂直攀爬的灵活性问题 locomotion
30 Design and Characterization of a Variable-Length Continuum Mechanism with Force Locking 提出一种新型连续机制以解决高强度操作中的刚度不足问题 manipulation
31 MR. POP: Multi-Robot Parallel Optimizing Planner for Almost-Surely Asymptotically Optimal Planning 提出MR. POP以解决多机器人运动规划中的全局最优路径问题 motion planning

🔬 支柱二：RL算法与架构 (RL & Architecture) (3 篇)

#题目一句话要点标签🔗⭐
32 Towards VLA-Dreamer: Refining VLA Behavior Using World Models 提出VLA-Dreamer以解决VLA模型数据需求高的问题 imitation learning world model world models
33 Learning Vision-Based Agile Gap Traversal: Differentiable Simulation with a Warm-Started Critic 提出两阶段强化学习框架以提高自主四旋翼穿越狭窄间隙的效率 reinforcement learning behavior cloning differentiable simulation
34 Can a Robot Read Braille? - Learning to Adapt Contact via Imitation Learning for Tactile Braille Recognition 提出自适应接触框架以解决机器人盲文识别问题 policy learning imitation learning

🔬 支柱九：具身大模型 (Embodied Foundation Models) (2 篇)

#题目一句话要点标签🔗⭐
35 Causeway: Restoring Task Accessibility for Instruction Switching in VLA Policies 提出Causeway以解决VLA策略中的指令切换问题 vision-language-action VLA
36 AuthGuard-R: Safety-Compliant Mission Hijacking and Dual-Gate Defense for LLM-Controlled Robots 提出AuthGuard-R以解决安全合规的任务劫持问题 large language model

🔬 支柱三：空间感知与语义 (Perception & Semantics) (1 篇)

#题目一句话要点标签🔗⭐
37 RECAST: From Log Replay to Closed-Loop Driving Simulation with View-Complete Actors 提出RECAST以解决动态驾驶模拟中的视图完整性问题 3D gaussian splatting gaussian splatting splatting ✅

⬅️ 返回 cs.RO 首页 · 🏠 返回主页